Conference Agenda

Lost (and found) in latent land: applications to weather and climate

Pierre Gentine

Columbia University, United States of America

Embracing the Tyranny of Testing

Moritz Hardt

Max Planck Institute for Intelligent Systems, Germany

Bridging Perception and Logic: An Abductive Learning Cycle for Semantically Anchored Facial Expression Recognition

Jens Johannes Gebele^1,2, Philipp Brune¹, Frank Schwab², Sebastian von Mammen²

1: Neu-Ulm University of Applied Sciences, Germany; 2: University of Würzburg

Human-in-the-loop Concept Discovery and Curation in Vision Foundation Models

Hyesu Lim¹, Jinho Choi², Jaegul Choo², Steffen Schneider¹

1: Helmholtz Munich, Germany; 2: KAIST AI, South Korea

TACTIC: Tabular-Attribute Conditioned Transformer for Image Classification

Laura Daza^1,2, Marta Hasny^1,2, Cristina González^1,2, Julia Schnabel^1,2,3

1: Helmholtz Munich, Germany; 2: Technical University of Munich, Germany; 3: King's College London, UK

Learning Physical Geometry from Noisy Helical Particle Tracks: A Comparative Study of Transformers, SBI, and JEPA

Sylvain Caillou, Jan Stark, Alexis Vallier, Catherine Biscarat

L2I Toulouse, CNRS/IN2P3, Université de Toulouse

MLflow pilot service for Helmholtz researchers

Lisana Berberi, Christophe Laures, Khadijeh Alibabaei, Valentin Kozlov, Achim Streit

Karlsruhe Institute of Technology (KIT), Germany

Towards Robust Foundation Models for Digital Pathology

Jonah Kömen^1,2, Edwin de Jong³, Julius Hense^1,2, Hannah Marienwald^1,2, Jonas Dippel^1,2,3, Philip Naumann^1,2, Eric Marcus⁴, Lukas Ruff³, Maximilian Alber^3,7, Jonas Teuwen⁴, Frederick Klauschen^1,5,6,7, Klaus-Robert Müller^1,2,8,9

1: Berlin Institute for the Foundations of Learning and Data (BIFOLD), Berlin, Germany; 2: Machine Learning Group, Technische Universität Berlin, Berlin, Germany; 3: Aignostics, Berlin, Germany; 4: The Netherlands Cancer Institute Amsterdam (NKI), Antoni van Leeuwenhoek Hospital (AvL), Amsterdam, Netherlands; 5: Institute of Pathology, Ludwig Maximilian University, Munich, Germany; 6: German Cancer Research Center, Heidelberg, and German Cancer Consortium, Munich, Germany; 7: Institute of Pathology, Charité Universitätsmedizin, Berlin, Germany; 8: Department of Artificial Intelligence, Korea University, Seoul, Korea; 9: Max-Planck Institute for Informatics, Saarbrücken, Germany

Sailing Past Syntax: A Human-in-the-Loop Framework for Safe Generative AI in Science

Nico Schuster

Centre de Physique des Particules de Marseille, France

The Helmholtz Model Zoo: Enabling AI Model Sharing and Inference in the Helmholtz Cloud

Hans Werners, Engin Eren, Philipp Heuser

Deutsches Elektronen-Synchroton DESY, Germany

Improving Reliability of LLM-Based Robotic Task Planning Through Domain Adaptation and Benchmarking

Kanak Pandit, Ananya Kulkarni, Muhammad Saeed

ARENA2036, Germany

Differentiable Wave-Optics for Single-Shot X-Ray Phase-Contrast Imaging of Plasma Targets

Ritz Ann Aguilar¹, Long Yang¹, Lingen Huang¹, Michael Bussmann^1,2, Ulrich Schramm^1,3, Thomas Cowan^1,4, Jeffrey Kelling^1,5

1: Helmholtz-Zentrum Dresden-Rossendorf, Germany; 2: Center for Advanced Systems Understanding, Germany; 3: Technische Universität Dresden, Germany; 4: European XFEL, Germany; 5: Technische Universität Chemnitz, Germany

Performance Bounds for Reliability and Hallucination Risk in Remote-Sensing Super-Resolution

David Iagaru^1,2, Jakob Gawlikowski¹

1: German Aerospace Center (DLR), Remote Sensing Technology Institute, Germany; 2: Ecole Polytechnique, Department of Applied Mathematics, Paris, France.

BioXPT-Brain: a foundation model integrating 3D vasculature and spatial transcriptomics to decode aging and vascular dementia

Laurent H. A. Simons^1,2, Louis B. Kümmerle^1,2,4, Fabian Theis^2,3,4, Ali Ertürk^1,5,6,7

1: Institute for Intelligent Biotechnologies, Helmholtz Zentrum München, Neuherberg, Germany; 2: Institute of Computational Biology, Helmholtz Zentrum München, Neuherberg, Germany; 3: School of Computing, Information and Technology, Technical University of Munich, Munich, Germany; 4: TUM School of Life Sciences Weihenstephan, Technical University of Munich, Munich, Germany; 5: Institute for Stroke and Dementia Research, Klinikum der Universität München, Ludwig-Maximilians University Munich, Munich, Germany; 6: School of Medicine, Koç University, İstanbul, Turkey; 7: Munich Cluster for Systems Neurology (SyNergy), Munich, Germany

Modelling Patient Variation Across Datasets And Diseases With Contrastive Learning On Single-Cell Data

Vladimir A. Shitov^1,4, Benjamin M. Frey^1,4, Mohammad Moghareh Dehkordi², Till Richter^1,2, Mojtaba Bahrami^1,3, Emma Maria Schonner², Youssef Naciri Zerta¹, Fabian J. Theis^1,2,3, Malte D. Luecken^1,4

1: Institute of Computational Biology, Computational Health Center, Helmholtz Munich, Munich, Germany; 2: School of Computing, Information and Technology, Technical University of Munich, Munich, Germany.; 3: TUM School of Life Sciences Weihenstephan, Technical University of Munich, Germany.; 4: Comprehensive Pneumology Center (CPC) with the CPC-M bioArchive and Institute of Lung Health and Immunity, Helmholtz Munich, Member of the German Center for Lung Research (DZL), Munich, Germany.

No Data? No Problem: Robust Vision-Tabular Learning with Missing Values

Marta Hasny^1,2, Laura Daza^1,2, Keno Bressem², Maxime Di Folco^1,2,3, Julia A. Schnabel^1,2,4

1: Helmholtz Munich, Germany; 2: Technical University of Munich, Germany; 3: Telecom Paris, France; 4: King's College London, UK

Reliable and Sustainable AI for Scientific Discovery

Gitta Kutyniok

LMU Munich, Germany

AI and Future of Medicine

Daniel Rueckert

Technical University of Munich, Germany

Deep Graph Learning for Temporal Data

Ingo Scholtes

University of Würzburg, Germany

Neurosymbolic Models of Uncertainty and Logical Reasoning

Annemarie Friedrich

Universität Augsburg, Germany

Rethinking the Foundations of Weather and Climate Modelling

Martin Schultz

FZJ, Germany

A Multi-Sensor Foundation Model for Earth Observation

Nassim Ait Ali Braham^1,2, Aaron Christian Banze^1,3, Julien Mairal⁴, Jocelyn Chanussot⁴, Conrad Martin Albrecht¹, Xiao Xiang Zhu²

1: German Aerospace Center (DLR), Germany; 2: Technical University of Munich (TUM), Germany; 3: University of the Bundeswehr Munich, Germany; 4: Universite Grenoble Alpes, Inria, CNRS, Grenoble INP, LJK, France

OneProtGPT: Bridging Protein Embeddings and Large Language Models for Protein Understanding

Javad Kasravi, Alina Bazarova, Stefan Kesselheim

Jülich Supercomputing Centre, Forschungszentrum Jülich, 52428 Jülich, Germany

AMPFormer: A Peptide Foundation Model for Antimicrobial Discovery

Pankhil Gawade^1,2, Adam Izdebski^1,2, Marcelo DT Torres⁴, Cesar De La Fuente-nunez⁴, Ewa Szczurek^1,3

1: Institute of AI for Health, Helmholtz Zentrum Munchen; 2: Technical University of Munich, TUM School of Computation, Information and Technology; 3: Faculty of Mathematics, Informatics and Mechanics, University of Warsaw; 4: University of Pennsylvania, Philadelphia, PA, USA.

The Mean is the Mirage: Entropy-Adaptive Model Mergingunder Heterogeneous Domain Shifts in Medical Imaging

Sameer Ambekar^1,2,3,4, Reza Nasirigerdeh^5,4, Peter J. Schuffler^5,1,4, Lina Felsner¹, Daniel M. Lang^1,2, Julia A. Schnabel^1,2,3,5,6

1: School of Computation, Information and Technology, Technical University of Munich, Germany; 2: Institute of Machine Learning in Biomedical Imaging, Helmholtz Munich, Germany; 3: relAI – Konrad Zuse School of Excellence in Reliable AI; 4: Munich Center for Machine Learning (MCML); 5: Institute of Pathology, Technical University of Munich, Germany; 6: School of Biomedical Engineering and Imaging Sciences, King’s College London, UK.

ConvexGating infers gating strategies from clusters in single cell cytometry data

Vincent D Friedrich^1,2, Karola Mai^3,4, Thomas P Hofer⁵, Elfriede Nößner⁵, Lorenzo Bonaguro^3,6,7, Celia L Hartmann^3,8, Aleksej Frolov^3,8,9, Caterina Carraro^3,6, Doaa Hamada^3,10,11, Mehrnoush Hadaddzadeh-Shakiba^3,8, Heidi Theis⁷, Dalila Juliana Silva Ribeiro¹², Dagmar Wachten¹², F Thomas Wunderlich¹³, Markus Scholz^1,14, Fabian J Theis^15,16,17, Matthias Becker^3,4, Marc D Beyer^3,7,8, Joachim L Schultze^3,6,7, Maren Büttner^3,7,15

1: University of Leipzig, Institute for Medical Informatics, Statistics, and Epidemiology, Leipzig, Germany; 2: Center for Scalable Data Analytics and Artificial Intelligence (ScaDS.AI), Leipzig, Germany; 3: Systems Medicine, Deutsches Zentrum für Neurodegenerative Erkrankungen (DZNE), Bonn, Germany; 4: Modular High Performance Computing and Artificial Intelligence, German Center for Neurodegenerative Diseases (DZNE), Bonn, Germany; 5: Research Group Tissue Control of Immunocytes, Helmholtz Center Munich, Munich, Germany; 6: Life and Medical Sciences (LIMES) Institute, University of Bonn, Bonn, Germany; 7: PRECISE Platform for Single Cell Genomics and Epigenomics, DZNE and University of Bonn and West German Genome Center (WGGC), Bonn, Germany; 8: Immunogenomics & Neurodegeneration, Deutsches Zentrum für Neurodegenerative Erkrankungen (DZNE), Bonn, Germany; 9: Department of Microbiology and Immunology, The Peter Doherty Institute for Infection and Immunity, University of Melbourne, Melbourne, VIC, Australia; 10: Molecular Immunology in Neurodegeneration, German Center for Neurodegenerative Diseases and the University of Bonn, Germany; 11: Medical Microbiology and Immunology Department, Faculty of Medicine, Mansoura University, Egypt; 12: Institute of Innate Immunity, Biophysical Imaging, Medical Faculty, University of Bonn, Bonn, Germany; 13: Max Planck Institute for Metabolism Research, Center for Endocrinology, Diabetes and Preventive Medicine (CEDP), Cologne, Germany; 14: University of Leipzig, Faculty of Mathematics and Computer Science, Leipzig, Germany; 15: Institute of Computational Biology, Helmholtz Center Munich, Germany; 16: Department of Mathematics, Technical University of Munich, Germany; 17: TUM School of Life Sciences Weihenstephan, Technical University of Munich, Germany

RenewBench: Real Energy Data You Can Actually Use

Elena Vollmer^1,3, Andreas Beckert^2,3, Hadeer El Ashhab^1,3, Paul Keil^2,3, Benjamin Schäfer^1,3, Beate Geyer^2,3, Markus Götz^1,3, Tobias Weigel^2,3, Kaleb Phipps¹

1: Karlsruhe Institute of Technology (KIT), Germany; 2: Helmholtz-Center Hereon, Germany; 3: Helmholtz AI, Germany

Cross Modalities Pretraining of Sparse Lidar and Dense Image Foundation Model for Global Carbon Stock Mapping

Audita Widya Astuti^1,2, Aldino Rizaldy¹, Gunjan Joshi¹, Weikang Yu¹, Peter Steinbach¹, Qi Zhang², Xiao Xiang Zhu², Pedram Ghamisi¹

1: Helmholtz-Zentrum Dresden-Rossendorf, Dresden, Germany; 2: Chair of Data Science in Earth Observation, Technical University of Munich, Munich, Germany

ICD-Code Extraction from Clinical Notes using Large Language Models in a RAG pipeline

Ludwig Krause¹, Martin Dyrba^1,2, Thomas Kirste¹, João Areias Saraiva^1,2

1: Hybrid Methods in Artificial Intelligence and Machine Learning, University of Rostock, Germany; 2: German Center for Neurodegenerative Diseases, Rostock, Germany

AI-enabled Colorimetric Multi-Biomarker Sensing Patch for Neonatal Monitoring

Alejandra Castelblanco^1,2, Elisabetta Ruggeri³, Giusy Matzeu³, Motaharehsadat Heydarian⁴, Kai Foerster^4,5, Ahmed Bahnasy², Andreas Flemmer⁵, Julia A. Schnabel^1,2,6, Benjamin Schubert¹, Fiorenzo G. Omenetto^3,8,9, Anne Hilgendorff^4,5,7

1: Computational Health Center, Helmholtz Center Munich, Neuherberg, Germany.; 2: School of Computation, Information and Technology, Technical University of Munich, Munich, Germany.; 3: Silklab, Dept. of Biomedical Engineering, Tufts University, Medford, MA, USA.; 4: Comprehensive Pneumology Center with the CPC-M bioArchive and Institute of Lung Health and Immunity, Helmholtz Center Munich, Member of the German Center of Lung Research (DZL), Munich, Germany; 5: Dr. von Hauner Children’s Hospital, Hospital of the Ludwig-Maximilians-Universität München, Munich, Germany; 6: School of Biomedical Engineering and Imaging Sciences, King’s College London, London, UK.; 7: School of Medicine and Health Sciences, Carl von Ossietzky University Oldenburg, Oldenburg, Germany; 8: Dept. of. Electrical and Computer Engineering, Tufts University, Medford, MA, USA.; 9: Dept. Of Physics, Tufts University, Medford, MA, USA.

Implicit Neural Representation (INR) meets Multi-Contrast MRI Reconstruction

Natascha Niessen^1,2, Carolin M. Pirkl², Ana Beatriz Solana², Hannah Eichhorn^1,3, Veronika Spieker^1,3, Wenqi Huang¹, Tim Sprenger^2,4, Marion I. Menzel^2,5,6, Julia A. Schnabel^1,3,7, Predictom Consortium²

1: School of Computation, Information and Technology, Technical University of Munich, Munich, Germany; 2: GE HealthCare, Munich; 3: Institute of Machine Learning in Biomedical Imaging, Helmholtz Munich; 4: Department of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden; 5: Technische Hochschule Ingolstadt, Ingolstadt, Germany; 6: School of Natural Sciences, Technical University of Munich, Munich, Germany; 7: King’s College London, London, United Kingdom

Deep Learning Reconstruction of Diffusion Spectrum Imaging from Undersampled q-Space Measurements

Roman Schmid¹, Sebastian Endt^1,2, Marion I. Menzel^1,2

1: Technische Hochschule Ingolstadt, Germany; 2: Technical University of Munich, Germany

HematoGraph: Graph-Aware Hierarchical Pooling for Cell-Level Hematology Classification

Furkan Dasdelen^1,3, Fatih Ozlugedik^1,2,3, Carsten Marr^1,3

1: Helmholtz Munich, Germany; 2: TUM; 3: LMU

Contour Proposal Networks with Deep Refinement for Dense High-Throughput Instance Segmentation

Eric Upschulte^1,2,3, Katrin Amunts^2,1, Timo Dickscheid^4,2,3

1: C. & O. Vogt Institute for Brain Research, University Hospital Düsseldorf, Germany; 2: Institute of Neuroscience and Medicine (INM-1), Research Center Jülich, Germany; 3: Helmholtz AI, Research Center Jülich, Germany; 4: Institute for Computational Visualistics, University of Koblenz, Germany

Video-Guided Policies for Robotic Manipulation

Cordelia Schmid

Inria, France

Helmholtz Munich: Discovering Future Health

Eleftheria Zeggini

Helmholtz Munich, Germany

Translational Genomics of Osteoarthritis

Eleftheria Zeggini

Helmholtz Munich, Germany

Reliable AI in the Age of Agentic Models

Leo Schwinn

Helmholtz Munich, Germany

Biophysics-Informed and AI-Assisted Computation for Translational Optoacoustic Imaging and Sensing

Dominik Jüstel^1,2

1: Helmholtz Munich, Germany; 2: Technical University of Munich, Germany

From Multivariate Perturbations to Predictive Models of Human Airway Function

Katzrin Murad, Benedetta Zampa, Weixu Wang, Inti De La Rosa Velázquez, Fabian Theis, Janna Nawroth

Helmholtz Zenter München GmbH, Germany

The Interaction Layer: Why the Virtual Cell Needs Biochemistry

Pascal Falter-Braun

Helmholtz Munich, Germany